The applicant's long term career-goal is to become a physician scientist dedicated to neuroscience. The ultimate direction is to develop improved methods of treating human epilepsy. The applicant plans to use a combination of gene transfer and neural grafting as a tool to study seizure propagation and as a potential therapeutic approach to epilepsy. Local application of GABA receptor agonist (e.g. muscimol) or GABA transaminase inhibitor (e.g. gamma-vinyl GABA) into discrete brain regions (e.g. substantia nigra, entopeduncular nucleus, amygdala and inferior colliculus) could suppress seizure in various animal models of epilepsy. Moreover, chronic infusion of GABA locally into motor cortex completely blocked the behavioral and electrographical seizure in naturally photosensitive baboon. The impermeability to blood-brain-barrier and global toxicity, however, preclude the systemic use of these agents as antiepileptic drugs. Therefore, the overall objective of this proposal is to test the idea that grafting the genetically modified, GABA-secreting fibroblasts into discrete brain regions can modify seizure manifestations. To address this objective, glutamic acid decarboxylase (GAD) gene will be used as the reporter gene, retroviruses will be the gene delivery vectors and both primary skin fibroblasts and fibroblasts cell lines will be the donor cells. The reasons for the use of fibroblasts as donor cells are (1) retroviruses require dividing cells for infection; (2) fibroblasts are readily available and easily cultured and manipulated in vitro; (3) essential to the long-term strategy of potential therapy is the use of primary skin fibroblasts as donor cells for autograft. Specifically, the proviral gene of wild-type retrovirus will be isolated to construct the plasmid by replacing the genes necessary for viral replication with the GAD gene. The transmissible but replication- incompetent retroviral vectors carrying GAD gene can be produced by introducing the constructed plasmid to the cells lines containing the proviral gene of retroviral helper virus. By transfection with the vectors, GAD gene will be integrated into fibroblasts. The in vitro expression of GAD gene will be quantitated by measuring mRNA and secreted GABA. To monitor the function in vivo, the genetically modified fibroblasts will be grafted into hippocampus or lateral ventricle and tested for the effects on the seizure phenomenon (behavioral and electroencphalographical seizure), seizure threshold and seizure-induced neuronal degeneration. The model systems for these purposes are a chronic model of epilepsy induced by fimbria-fornix lesions and an acute model of limbic seizure induced by intraamygdala injection of kainic acid. The long-term graft survival, in vitro gene expression and host immune reaction will be assessed by immunohistochemical approaches using antibodies to various specific cell markers.